INDEX

Explanations

names related to political figures and terms

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pires

-0.64

aughtered

-0.61

emonium

-0.61

©¶æ

-0.60

aughed

-0.59

iless

-0.59

 Airl

-0.58

Jol

-0.57

 Airbus

-0.56

thens

-0.54

POSITIVE LOGITS

Semitism

0.86

 versus

0.83

 being

0.83

 involvement

0.83

 happening

0.78

anship

0.77

being

0.76

 lately

0.76

liness

0.75

 disappearing

0.75

Activations Density 5.598%

names related to political figures and terms

No Comments

No Known Activations